Performance of Immunoglobulin G Serology on Finger Prick Capillary Dried Blood Spot Samples to Measure SARS-CoV-2 Humoral Immunogenicity

Importance

Measuring humoral immunogenicity of Severe Acute Respiratory Syndrome Coronavirus 2 vaccines and finding population-level correlates of protection against coronavirus disease presents an immediate challenge to public health practitioners.

Objective

To study the diagnostic accuracy and predictive value of finger prick capillary dried blood spot samples tested using an anti-immunoglobulin G (IgG) serology assay to measure SARS-CoV-2 seropositivity and the humoral immunogenicity of COVID-19 vaccination.

Design, Setting and Participants

This cross-sectional study enrolled participants (n= 644) who had paired DBS and serum samples collected by finger prick and venipuncture, respectively, in British Columbia, Canada between January 12 ^th , 2020 and May 21 ^st , 2021. Samples were tested by a multiplex electrochemiluminescence assay for SARS-CoV-2 anti-Spike (S), -Nucleocapsid (N) and -receptor binding domain (RBD) IgG reactivity using a Meso Scale Discovery (MSD) platform. Additionally, unpaired DBS samples (n= 6,706) that were collected in the province during the same time period were included for analysis of SARS-CoV-2 anti-N IgG reactivity.

Exposure

Collection of a capillary dried blood spot by finger prick alone or paired with serum by venipuncture.

Outcome

Humoral immune response to SARS-CoV-2 measured by detection of anti-S, -N or - RBD IgG.

Results

In comparison to a paired-serum reference, dried blood spot samples possess a sensitivity of 80% (95% CI: 61%-91%) and specificity of 97% (95% CI: 95%-98%). Receiver operator characteristic curve analysis (ROC) found that participant DBS samples tested for anti-SARS-CoV-2 IgG by MSD V-PLEX COVID-19 Coronavirus Panel 2 assay accurately classify SARS-CoV-2 seroconversion at an 88% percent rate, AUC= 88% (95% CI: 81%-96%). Modelling found that a dried blood spot-based testing approach has a high positive predictive value (98% [95% CI: 98%-99%]) in a theoretical population with seventy-five percent COVID-19 vaccine coverage. At lower vaccine coverages of fifteen and forty-five percent, the test’s positive predictive value decreased, and the negative predictive value increased.

Conclusion

We demonstrate that dried blood spot collected samples, when tested using an electrochemiluminescence assay, provide a valid alternative to traditional venipuncture and should be considered to reliably detect SARS-CoV-2 seropositivity.

Key Points